The assembly of human herpesviruses, specifically the envelopment of the virion, has remained an active by contentious area of investigation. Two different pathways of assembly have been proposed for alpha-herpesviruses based on studies in HSV, VSV, and PRV. Although considerable evidence has suggested that envelopment occurs at the nuclear membrane, more recent findings have argued for cytoplasmic envelopment following an envelopment/deenvelopment step at the nuclear envelope. The assembly of beta-herpesviruses, specifically human cytomegalovirus (HCMV), appears to differ from that of alpha-herpesviruses in that the final tegumentation and compartment in which both tegument and envelope proteins accumulate and have suggested this is a cytoplasmic site of assembly. Furthermore, the PI and others have shown that tegument coated particles bud into cytoplasmic vacuoles whose limiting membranes contain viral envelope glycoproteins. More recently, the PI has described biochemical and imaging findings which indicate that the cytoplasmic tail of the major envelope glycoprotein of HCMV, gB, interacted with an abundant, myristylated tegument protein, pp28. Together these findings have argued for an assembly pathway which is similar to that described for RNA viruses. This pathway included accumulation of structural proteins in an cytoplasmic compartment, specific interactions between matrix (tegument) proteins and envelope glycoprotein, deformation of host membranes and exclusion of host proteins, and finally, budding of the particle. In this proposal, the PI will investigate this model by initially characterizing the interaction between gB and pp28 and between other viral proteins, which interact with these two structural proteins. They will then define targeting signals, which direct these proteins to cytoplasmic sties of assembly. In the final section of the proposal, they will generate recombinant viruses with mutations in gB, pp28 and interacting proteins which based on earlier finding, will interrupt or interfere with the assembly program. Analysis of the phenotypes of these viruses will all them to directly address their hypothesis and determine the specific protein interactions between tegument proteins and envelope glycoproteins which are crucial to the assembly of HCMV.